Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.33 as permitted sender) client-ip=23.128.96.33;
Date:   Fri, 15 Sep 2023 10:32:05 -0700
In-Reply-To: <8b047dad-84ac-69f9-3875-38bca92d7534@amd.com>
Mime-Version: 1.0
References: <cover.1694721045.git.thomas.lendacky@amd.com> <8a5c1d2637475c7fb9657cdd6cb0e86f2bb3bab6.1694721045.git.thomas.lendacky@amd.com>
 <ZQNs7uo8F62XQawJ@google.com> <f2c0907c-9e30-e01b-7d65-a20e6be4bf49@amd.com> <8b047dad-84ac-69f9-3875-38bca92d7534@amd.com>
Message-ID: <ZQSVFQ78M/OUtWaj@google.com>
Subject: Re: [PATCH 1/2] KVM: SVM: Fix TSC_AUX virtualization setup
From:   Sean Christopherson <seanjc@google.com>
To:     Tom Lendacky <thomas.lendacky@amd.com>
Cc:     kvm@vger.kernel.org, linux-kernel@vger.kernel.org, x86@kernel.org,
        Paolo Bonzini <pbonzini@redhat.com>,
        Borislav Petkov <bp@alien8.de>,
        Dave Hansen <dave.hansen@linux.intel.com>,
        Ingo Molnar <mingo@redhat.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Babu Moger <babu.moger@amd.com>
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Fri, Sep 15, 2023, Tom Lendacky wrote:
> On 9/14/23 15:48, Tom Lendacky wrote:
> > On 9/14/23 15:28, Sean Christopherson wrote:
> > > On Thu, Sep 14, 2023, Tom Lendacky wrote:
>=20
> >=20
> > >=20
> > > > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 if (guest_cpuid_has(vcp=
u, X86_FEATURE_RDTSCP))
> > > > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=
 svm_clr_intercept(svm, INTERCEPT_RDTSCP);
> > >=20
> > > Same thing here.
> >=20
> > Will do.
>=20
> For RDTSCP, svm_recalc_instruction_intercepts() will set/clear the RDTSCP
> intercept as part of the svm_vcpu_set_after_cpuid() path, but it will onl=
y
> do it based on kvm_cpu_cap_has(X86_FEATURE_RDTSCP) being true, which is v=
ery
> likely.
>=20
> Do you think that is good enough and we can drop the setting and clearing=
 of
> the RDTSCP intercept in the sev_es_vcpu_set_after_cpuid() function and on=
ly
> deal with the TSC_AUX MSR intercept?

The common handling should be good enough.

> On a side note, it looks like RDTSCP would not be intercepted if the KVM =
cap
> X86_FEATURE_RDTSCP feature is cleared, however unlikely, in
> kvm_set_cpu_caps() and RDTSCP is not advertised to the guest (assuming th=
e
> guest is ignoring the RDTSCP CPUID bit).

Hmm, yes, though the only scenario in which KVM clears RDTSCP on AMD comes =
with
a WARN (it's a guard against KVM bugs).  If the guest ignores CPUID and use=
s
RDTSCP anyways, the guest deserves its death, and leaking the host pCPU doe=
sn't
seem like a major issue.

That said, if hardware behavior is to ignore unknown intercepts, e.g. if KV=
M can
safely set INTERCEPT_RDTSCP even when hardware doesn't support said interce=
pt,
then I wouldn't be opposed to doing:

	/*
	 * Intercept INVPCID if shadow paging is enabled to sync/free shadow
	 * roots, or if INVPCID is disabled in the guest to inject #UD.
	 */
	if (!kvm_cpu_cap_has(X86_FEATURE_INVPCID) ||
	    !npt_enabled || !guest_cpuid_has(&svm->vcpu, X86_FEATURE_INVPCID))
		svm_set_intercept(svm, INTERCEPT_INVPCID);
	else
		svm_clr_intercept(svm, INTERCEPT_INVPCID);

	if (kvm_cpu_cap_has(X86_FEATURE_RDTSCP) &&
	    guest_cpuid_has(vcpu, X86_FEATURE_RDTSCP))
		svm_clr_intercept(svm, INTERCEPT_RDTSCP);
	else
		svm_set_intercept(svm, INTERCEPT_RDTSCP);

Alternatively, KVM could check boot_cpu_has() instead or kvm_cpu_cap_has(),=
 but
that's not foolproof either, e.g. see Intel's of hiding PCID to workaround =
the
TLB flushing bug on Alderlake.  So my vote would either be to keep things a=
s-is,
or do the above (if that's safe).